Guilty Gear Strive: The Gear That Never Will Be

I was still so stubbornly holding out hope that this wouldn’t be a game designed from the ground up to cause clinical depression in its players. This is the true origin point of my hater arc, though I was still in denial.

That said, I can certainly tell you who this game was for: people who wanted a sequel to Street Fighter V. Whoops!

I think what we can take away from this is that one key point — the primary focus of Guilty Gear 2020’s design is comprehensibility. This could fundamentally change what Guilty Gear is, to be sure, but by that same token, it could be a chance for Guilty Gear to show us how it’ll handle its world-famous brand of craziness with a more focused design philosophy.

“Do we need a new world order? – Fighting Game Accessibility and Guilty Gear 2020”

Late 2019 was a simpler time.

Among all the announcements that came out of that year’s Evo (the FGC’s favourite trade show masquerading as a major tournament), one had people particularly excited — a new Guilty Gear was in the works. The most recent game in the series at the time, Guilty Gear Xrd, was pushing five years old, and with two major revisions under its belt in Revelator and Rev2, people were excited to see what would become of the series moving into the new decade. The trailer, in usual Arc System Works fashion, looked pretty and said as little as possible. Featuring the developer’s now instantly recognisable 3D-2D art direction, some interesting interpolation effects, a few crazy cinematic camera angles and a memorable (if maybe for the wrong reasons) new title track from now-mainstay series musician, Naoki Hashimoto, the trailer told everyone one thing in particular:
“We already know the smell of the game.”

I’m not so sure they do.

ArcSys ran a closed beta test for Guilty Gear Strive over the weekend of the 17th of April. Many who were able to play the game at FGC majors and trade shows got a guaranteed place in the beta as long as they filled out the company’s survey — the rest had to play the lottery. I was lucky enough to get a code for the game and a link to the very same survey. I spent as much time as I could playing the game, both against the CPU and matching against people in the online lobbies. I’ve covered (and will probably continue to cover) my thoughts on the game itself pretty extensively on my Twitter account, so I’ll just give you the cliff notes on all of those thoughts.

The online lobbies are dogshit and I really don’t understand what was wrong with Rev2‘s lobbies or even just a fucking menu. I wholly expect the game to be delayed just so that the lobbies can be reworked, because I gotta tell ya, playing footsies in some unholy union of Habbo Hotel and MapleStory just to play a match is not it, and we haven’t even gotten into the exceptionally poor way the game handles sorting players into lobbies based on their skill level.

The UI would honestly be fine if not for the fact that the Burst and RISC meters are crammed into the tiny space provided by the game’s moving portraits. If these elements were made more prominent and given their own static space on the HUD then that’d be all you’d really need to change from a functional standpoint. The combo counter would also be perfectly fine if it didn’t increase in size beyond where it gets to in the 5-8 hit range, but instead Faust lands one command grab and we suddenly have a combo counter that covers half the screen and looks god-awful on top of obscuring the action to a heinous degree.

Neutral feels good and still more or less like Guilty Gear, but offense feels incredibly stilted thanks to the significantly limited gatling chains, which also makes pressuring with every character feel really homogeneous, not to mention inflexible. An expansion of the gatlings is something that a few players have requested for some time now, but even just allowing P-to-K gatlings would go a long way to help with fluidity.
The new forward air dash makes sense from a design standpoint, but the roughly 12 frames of startup makes using it in neutral feel incredibly sluggish, even if the ability to buffer attacks or Faultless Defense during startup significantly increases its offensive potential (and also makes it really fucking hard to anti-air), especially thanks to the new ability for certain air normals to be cancelled directly into an air dash. I think you could get away with lowering the startup to something like 3-5 frames as well as keeping the DMC-style seal effect, since that way we’re basically dealing with a Melty Blood air dash, but if we went back to just the old air dashes I wouldn’t complain.

Defense feels less dynamic and less interactive as well, largely thanks to the changes surrounding air blocking and Instant Block. You can now air block everything, but blocking stuff in the air incurs much heavier RISC generation. I don’t think this makes much sense for the current system, since being in the air is already exceptionally strong, and now most players can just up-back against every pressure sequence and have very little to worry about outside of a throw or maybe getting caught out during jump startup. Even just giving characters a few attacks that were designated anti-air attacks would help significantly. I mean, seriously, seeing people block Potemkin’s Heat Knuckle just isn’t right.
As for Instant Blocking, instead of decreased blockstun, a successful IB builds a significant amount of Tension. There’s at least a clear design rationale here, in that the game seems to be placing a very heavy emphasis on resource management in general, and the restricted gatling system creates less of a need to find gaps in pressure since they’ll be occurring naturally more frequently anyway. The problem, though, is that there are so many other ways to quickly build meter in this game that doing so through IB feels kind of redundant, and it leaves many people feeling as though IB as a mechanic just doesn’t exist, leaving them only with Faultless Defense, which while undoubtedly useful, is essentially pulling double duty in a much less interesting way than before.

The new mechanics surrounding wall interactions in the corner are also really obtuse and disruptive in some ways. There seem to be very few restrictions on what attacks are allowed to interact with the wall, and while the more freeform corner combos that this facilitates do look really cool and feel great to do, the fact that wall damage regenerates at a static rate with no visual indicator as to its status means that reward at the wall feels inconsistent, sometimes to the point of feeling like a punishment for choosing okizeme over the extra damage and meter gain that comes from a stage transition. There are a few ways that you can go about fixing this, the easiest of which being making wall damage regenerate faster (or just reset entirely after a wall combo ends) and allowing players to take a knockdown after getting a wall stick rather than forcing the wall break, but this whole dynamic could be made much more elegant by adding more restrictions on what moves cause wall bound and wall stick, as well as changes to wall damage I’ve already outlined. Right now, the wall mechanics feel like something that players shouldn’t really even want to be interacting with if they can help it, which doesn’t exactly signal an interesting dynamic.

We also need to talk about RISC, as well as combo damage. The simplistic and homogeneous pressure afforded by the reworked gatling chains feels especially egregious when you can build up your opponent’s RISC gauge to nearly 50% just by making them block 5P three times, and combined with the quite literally preposterous damage you often get from most combos (or even some individual moves — Sol’s Wild Throw stands out as being a command grab on a very mobile character that deals roughly 35% damage on average), it makes the risk/reward in most offensive situations totally jacked, which certainly doesn’t help with how bad defense feels in this game. It also makes rounds end so fast that Leffen would absolutely be calling for this game to be played best of 7 in tournament if he ever got the chance.

All of that being said, there are some things to like. Aside from some suspect lighting in a few parts of some stages, the game is absolutely gorgeous. Throws being moved from the H button to the D button is something that warrants further exploration (especially given the slower startup and new whiff animation), but I don’t dislike it. It still seems reasonably good as a defensive tool, and its new functionality also makes it much easier to use on offense. I also like how the new input removes the old throw option selects, as well as prevents characters with stupid j.H attacks from option-selecting air throws with what is often one of their best aerial approach tools. The new juggle-centric combo system also feels very fun and interesting to play around with, and the new Roman Cancel system is arguably where the game has iterated and innovated the most. It’s had so much thought put into it that I kind of want ArcSys to just make it the focal point of the entire game — y’know, kind of like Xrd, except on purpose.
There are now four types of Roman Cancels: Red RC cancels offensive actions upon contact, Purple RC cancels offensive actions during their recovery, Blue RC is performed while no offensive action is being taken, and Yellow RC functions as the new universal guard cancel mechanic, replacing Dead Angle Attacks. There are also three new wrinkles to the system — firstly, RC slowdown still exists, but it now requires you to be close to your opponent, where upon activation it will apply a slowdown debuff to them (signified by a clock icon above their head), which very elegantly eliminates most of Xrd’s trademark “pause-button” neutral interactions that were facilitated by that game’s Yellow Roman Cancel. Next, the game allows you to cancel the RC animation into a special move, bypassing the usual “time stop” portion of the activation in a motion the game refers to as “Fast RC”, allowing different kinds of pressure sequences that deny your opponent much time or space to react to what’s happening. Finally, it is now possible to buffer a dash during the “time stop” portion of the RC activation in order to change your character’s momentum, referred to as “RC Drift”. Not only does this allow some interesting new combo extensions, it also has been used to set up very ambiguous left/right mixups when performed directly over an opponent’s head. The new RC system has a lot of depth and complexity to it, and I really look forward to playing around with it more.

I’m also very excited about the fact that this game will be using rollback netcode.

And that’s basically everything I’ve said on the game, both to ArcSys and to the void that is Twitter. Now I can start getting to my point.

I have no idea what this game wants to be.

The last time I spoke about this game, back when it was still only known as Guilty Gear 2020, I theorised that Daisuke Ishiwatari and Akira Katano, through all their talk of “beginner mechanics” and whatnot, were looking to make a game that was not so much easy to play, but easy to understand. I thought that this new Guilty Gear game, being approached from the ground up by the developers, would be a chance to show everyone what Guilty Gear’s brand of crazy would look like with a more focused design philosophy. What I was shown throughout this closed beta test was neither.

A cluttered and difficult-to-read user interface is the least of our problems. The old gatling system, while not without its own rules and nuances, could largely be summarised as “SMALL BUTTON GO TO BIG BUTTON”. This is something you don’t even have to explain to people, and yet the “easier to understand” version of this mechanic apparently looks like segmenting attack buttons into groups with specific chaining rules that seem more like the long-form target combos of Real Bout Fatal Fury 2 than a universal chain combo system. The game apparently also has gatling chains that are exclusive to counter hits, except no one knows if it’s actually that or if the game just lets you do whatever you want on counter hit as if it were a clash. The wall mechanics seem to require a PhD-level dissertation in order to fully understand how they work and what people are supposed to do with them. Air dashes now have a new means of altering their momentum, and while the utility is certainly interesting, that utility is now attached to an aerial mobility option almost doesn’t look like it’s doing what it’s supposed to do. Nearly every hit the opponent lands on you results in damage second only to Ultra Fight Da! Kyanta 2, except for when you’re on your last 25% and you effectively get a second life bar. And look, I understand that the beta won’t have a full-fledged tutorial. But if you’re going to make a series of videos telling me about all of your characters and which ones have a DP that I should be mashing on wakeup, then the very least you can do is make a video that also explains the mechanics behind the reworked (and now highly technical) Roman Cancel system.

It’s clear that accessibility is not the goal. By all accounts, Guilty Gear Strive is less intuitive than any previous game in the series. As it stands, I certainly wouldn’t be recommending this as a gateway into Guilty Gear, and not just because I’m a ride-or-die Accent Core +R player. I just cannot see new players being able to intuitively make sense of what is happening in this game.

So if the game isn’t actually for new players, who is it for? I can’t tell you that, either. A lot of this game’s design decisions, while in many cases interesting and seemingly following a reasonable internal design logic, tend to sacrifice good game feel on top of not really even justifying their existence, let alone contextualising any of the other mechanical changes. The inactionable period on forward air dashes has effectively been moved from during the movement to before it, which grants it new offensive utility, but only at the expense of making players feel strangely immobile. Restricted gatlings serve the purpose of shortening block strings and preventing big combos from low-commitment buttons (I guess combo scaling is too complicated?), but it results in extremely basic, homogenised pressure that feels unnecessarily inflexible. The wall mechanics serve to prevent people from being locked down in the corner forever, but the forced reset to neutral that comes with any stage transition not triggered with a super leads to a lot of players feeling as though the game is punishing them for playing correctly. Anti-airs forcing increased RISC gain as opposed to just swatting people out of the skies serves to reinforce the game’s attempt at pushing its idea of Guilty Gear offense, but it results in people blocking what is clearly meant to be a grab, which just feels all sorts of wrong. Instant Block only granting meter seems to make sense when you consider the limited gatlings and the new YRC (Roman Guard Cancel?), but the game hasn’t significantly weakened offense in any other respect, and the plethora of ways you can more reliably gain meter makes IB feel like a dead mechanic. It all makes sense on paper, but none of it feels good in practice.

The result is a game that feels very messy and confused. It is still very much in the “this game has potential” phase in my eyes, but only because the game is 7-8 months away at minimum. If ArcSys wants to make good on that potential, they need to really figure out what they want this game to be, lest it release as a slightly more polished version of the vaguely Guilty Gear-shaped blob we have right now.

However, irrespective of the direction the game takes, I don’t think it’s going to be in line with the Guilty Gear people have come to know and love over the past few years, and even decades.
The ways in which this game has broken with series tradition are not so easily walked back on. It can bring back all of the legacy systems people want, but ultimately, Guilty Gear Strive will not be anything other than, well, Guilty Gear Strive.

Guilty Gear Strive is, if anything, representative of how Arc System Works apparently sees Guilty Gear as a whole — Guilty Gear is the world that Sol, Ky, May and others inhabit, rather than the specific collection of systems and mechanics that many of its players see it as. I still personally have hope for the game, and if ArcSys really is taking survey feedback seriously, then I think it will genuinely be a great fighting game.
But all we can do is see it, play it, critique it, for what it is. Because it’s not going to be Guilty Gear as we know it. And for better or worse, it probably never will be.